Proc. Natd. Acad. Sci. USA Vol. 89, pp. 5872-5876, July 1992 Biochemistry

Design and synthesis of a CD4 a-turn mimetic that inhibits human immunodeficiency virus envelope glycoprotein gpl20 binding and infection of human lymphocytes (AIDS/molecular mdeg)

SHAOXING CHEN*, R. ALAN CHRUSCIEL*, HIROSHI NAKANISHIt, APICHAYA RAKTABUTR*, MICHAEL E. JOHNSONt, ALICE SATOt, DAVID WEINERt, JIM HOXIE§, HORACIO URI SARAGOVII, MARK I. GREENE¶, AND MICHAEL KAHN* II Departments of *Chemistry (rn/c 111), Box 4348, and tMedicinal Chemistry and Pharmacognosy (m/c 781), Box 6998, University of Illinois, Chicago, IL 60680; tWistar Institute, Thirty Sixth Street at Spruce, Philadelphia, PA 19104-4268; and Departments of §Medicine and Pathology and Laboratory Medicine, School of Medicine, University of Pennsylvania, Philadelphia, PA 19104-6082

Communicated by Harry H. Wasserman, March 23, 1992 (received for review January 29, 1992)

connecting the C' and C" (3-strands. In addition, based on mutagenesis experiments, a significant role for binding to gp120 has been attributed to Phe-43. However, peptides derived from this loop region have failed to exhibit significant inhibitory activity (12, 16). This is presumably due to a loss of critical secondary structure elements in the peptides and highlights the necessity for developing conformationally constrained compounds (17). Peptides are characteristically highly flexible molecules whose structure is strongly influenced by their environment (18), and their random conformation in solution may preclude their practical application for the study of molecular recognition processes (19). We have developed a method to construct conformationally restricted mimetics of peptidechain reversals (J3-turns, y-turns, fl-loops, etc.) (20). The surface localization of turns in proteins, and the predominance of residues containing potentially critical pharmacophoric information, has led to the belief that turns play important roles in a myriad of molecular recognition events (21). Our turn mimetic (Fig. la) possesses considerable potential for delineating structure-function relationships (20). The general system allows for the facile introduction of any natural or unnatural amino acid side-chain functionality through a modular component synthesis. The ability to alter the X-group linker provides reliable control and variation of side-chain orientations, backbone distances, and Oq dihedral angles. We have determined by molecular modeling that the 10membered ring mimetic framework (X = NH) (Fig. la) closely mimics the (3-turn at Gln4Q-Phe43. Based on this analysis, we have synthesized the conformationally restricted peptidomimetic 1 (Fig. lb), which incorporates CD4 residues 40-45 by the protocol outlined in Fig. 2. An overlay of the mimetic structure with that of CD4 is shown in Fig. 3. This first generation small molecule mimetic (Mr 810, as its trifluoroacetate salt) abrogates the binding of human T-lymphotropic virus type IIIB gpl20 to CD4+ cells at low micromolar levels and reduces syncytium formation 50%o at 250

Poor bioavailability, rapid degradation, anABSTRACT tigenicity, and high cost often limit the use of proteinaceous pharmaceuticals. One goal of structural biochemistry is the

reduction of complex molecules to small functional units that are amenable to high-resolution structural analysis and rapid modification. The dissection of complex proteins into small synthetic conformationally restricted components is an important step in the design of low molecular weight nonpeptides that mimic the activity of the native protein. We have developed a reverse-turn mimetic system to explore peptide and protein structure-function relationships. We now report the design and synthesis of a small molecule (Mr 810, as its trifluoroacetate salt), water soluble, proteolyticaly stable mimetic of residues Gln4O-Thr45 of the complementarity-determining 2-like region of CD4. This mimetic has a low micromolar Kd for human T-lymphotropic virus type hUB gpl20 and reduces syncytium formation.

CD4 is a cell-surface glycoprotein that is found principally on T lymphocytes (1), which associates with class II major histocompatibiity molecules on antigen-presenting cells (2, 3). CD4, via high-affinity binding (Kd =10-9 M) to the human immunodeficiency virus (HIV) viral envelope glycoprotein gpl20, also serves as a cellular attachment site for HIV (4, 5), although other binding sites may also be important (6). Several recombinant CD4-derived candidates have been developed as anti-HIV therapeutics (7). In general, poor bioavailability, rapid degradation, and antigenicity affect the utility of proteinaceous pharmaceuticals (8, 9). The dissection of complex proteins into small synthetic conformationally restricted functional subunits may overcome these problems. Mimetics of important functional domains might possess beneficial properties in comparison to the intact proteinaceous species with regard to specificity and therapeutic potential, and they should be valuable probes for the study of molecular recognition events. Extensive mutagenesis (10, 11) and peptide mapping experiments (12, 13) have been used to evaluate critical interactions involved in binding CD4 to gp120. There exists some deviation in pinpointing the exact range of the contact surface. However, a single stretch of amino acid residues 40-55 within the complementarity-determining 2-like region has been implicated as being essential. The x-ray structure of an N-terminal 182-residue fragment of CD4 was recently determined (14, 15). Inspection of the structure localizes residues Gln4Q-Phe43 to a highly surface-exposed (-turn

.ug/ml. MATERIALS AND METHODS The (-turn mimetic 1 was synthesized according to Fig. 2. Azetidinone 2 was prepared in six steps (22% yield) from Abbreviations: HIV, human immunodeficiency virus; F-gp120, fluorescein-conjugated gp120; sCD4, soluble CD4. ITo whom reprint requests should be addressed at present address: Department of Pathobiology, F161 Health Science SC38, University of Washington, Seattle, WA 98195.

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact. 5872

Proc. Natl. Acad. Sci. USA 89 (1992)

Biochemistry: Chen et al.

300,000 CD41 cells (from a stock of 10 x 107 cells per ml) were added to tubes at 40C in binding buffer, with a final vol of 100 jl. Samples were incubated at 40C for 40 min, washed in binding buffer, and analyzed in a fluorescence-activated cell sorter immediately. Reduction in fluorescence intensity was used to monitor the inhibition of F-gpl20 binding to CD4+ cells. Data were acquired gating on live cell populations (always >90%) and were consistent whether mimetics, F-gp120, or other agents were or were not added. The .-turn mimetic CD4-1 (Mr 810, as its trifluoroacetate salt) was soluble in water at 2-4 mg/ml. The Kd of CD4 mimetic 1 was estimated by the ChengPrusoff equation (29)

a

NHBn

b 0

5873

OH

Kd (CD4-1) =

IC50 (CD4-1) 1 + {[F-gpl20]/Kd(F-gpl20)}

H

HN

NH

N H

The Kd of F-gp120 was determined from Scatchard analysis to be 22 nM, which is consistent with the published data (30). The IC50 measurements were performed five times. The IC50 values obtained were between 6.1 and 30.6 ,uM. The CD4-1 mimetic was used for inhibition of syncytia formation. SupT1 cells were used as target cells for infection (6). Dilutions (1:2) of soluble CD4 (sCD4), CD4-1 mimetic, or the CD4 hexapeptide (amino acids 40-45) were made in 96-well plates in RPMI 1640 medium containing 10% fetal calf serum. H9/IIIB cells were then plated at a density of =104 cells per well. SupT1 target cells were added (5 x 105 cells per well) and syncytium formation was qualitatively and quantitatively determined after a 3-day incubation period.

FIG. 1. (a) Generic structure of mimetic ring system. X = 1-5 atoms. (b) CD4 loop region of peptidomimetic 1.

dimethyl-D-aspartate in a fashion analogous to procedures described by Salzmann et al. (23) and Williams et al. (24). Mixed anhydride coupling of 2 to O-benzyl serine benzyl ester and subsequent hydrogenolytic cleavage of the benzyl ester provided acid 4 in 88% yield. Coupling of the Z-protected hydrazinophenylalanine 5 (25) to 4 by the procedure of Tung et al. (26) proceeded smoothly. Reductive closure (27) and saponification afforded the 10-membered ring /-turn mimetic 6 in 53% yield. Dipeptide 7 was coupled to 6 by an aqueous carbodiimide protocol. Hydrogenolytic deprotection of the side-chain protecting groups and purification by reverse-phase C18 HPLC provided homogeneous material, which was characterized by 400-MHz NMR and fast atom bombardment MS. The NMR studies were carried out on a Bruker AM400 spectrometer in either C2HC13 or C2H302H at

250C. Minimum energy conformers of the mimetic were found by a Monte Carlo search with the Macromodel program BATCHMIN (22). A modified Macromodel MM2 force field, in which the N(sp2)-N(sp2) force constant was adapted from the Amber force field, was used for the calculations. There are two low-energy conformers, with the principal structural

difference being reversal of the peptide bond orientation between residues 2 and 3. The energy difference is -0.3 kcal/mol (1 cal = 4.184 J) in vacuum; molecular dynamics simulation in a volume-based continuum solution model (28) indicates that there is no significant energy difference between the two conformers under simulated solution conditions. Binding assays for inhibition of gpl20 binding to CD4+ cells, using reduction in fluorescence intensity upon binding, were performed as follows: fluorescein-conjugated gp120 (F-gpl2O) (F/P ratio 4.267) was incubated with the mimetics or with soluble CD4 at 22°C in binding buffer (Ca2+-, Mg2+free Hanks' balanced salt solution/0.5% bovine serum albumin/0.05% sodium azide, pH 7.4, for 40 min). Approximately

RESULTS AND DISCUSSION Mimetic 1 was assayed for its ability to inhibit binding of fluoresceinated human T-lymphotropic virus type IIIB gpl20 (F-gpl20) (30) to CD4+ Jurkat and SupTi human cells (Fig. 4; see ref. 8). The compound competitively inhibits F-gpl20 binding to CD4 in a concentration-dependent manner with an apparent Kd in the low micromolar range.** In contrast, our previously described reovirus receptor-specific 10-membered ring reverse-turn mimetic (YSGSS) (20) and the linear hexapeptide Gln4--Thr45 had no effect on the binding of F-gpl20. The failure of the reovirus receptor mimetic to inhibit gpl20 binding stresses the importance of side-chain orientation in mediating binding, since the two frameworks are identical. These studies illustrate the critical role that the P-turn framework plays in appositely positioning pharmacophoric information in a manner comparable to its presentation by the intact CD4 molecule. Syncytium formation is effected by the interaction of gpl20, expressed in the membranes of infected cells, with the CD4 receptor on neighboring cells (31). We examined the ability of the CD4-1 mimetic to inhibit gpl20-CD4-dependent fusion processes in a cocultivation model. At a concentration of 250 ,g/ml, there is an 50%o decrease in the number of syncytia formed (Fig. 5). It has been suggested that binding of sCD4 to F-gpl20 is temperature dependent, and that the affinity of sCD4 for F-gpl20 at 4°C decreases by 1 order of magnitude when compared to binding at 37°C (32). We found that the CD4-1 mimetic exhibits similar behavior and most efficiently inhibited F-gpl20 binding to cells when preincubated at 37°C (data not shown). It is interesting that the mimetic of the compleand **The Kd for the CD4 mimetic 1 was estimated to be between 4were 20 A.M based on the Cheng-Prusoff equation. These values corroborated by a standard commercially available ELISA (American Biotechnologies, Cambridge, MA).

5874

Biochemistry: Chen et al.

Proc. Natl. Acad. Sci. USA 89 (1992) I) iBuOCOCI, NMM 0

2)

TlIMMSO

2)

0III

HH5l CO2Me

ZNN

H

CO2Bn

H2N

5

3 2

3) H2, Pd/C, MeOH

4

3) H2, Pd/C, EtOAc

4) K2CO3, Aq. MeOH NHBn 04

OH

"1

HOK\ 1)

H N H2N

NHTBDMS °

0

7 H

CONHBn

NH

Y

-

"'OBn H

EDC, HOBT, Aq. THF

Il

=0

2) H2, Pd/C, MeOH satd. HCO OBn 3)

C8 HPLC

H

6

1

FIG. 2. Synthesis of 13-turn mimetic 1. TBDMS, tertiary butyl dimethyl silyl; TFA, trifluoroacetic acid; EDC, 1-ethyl-3(3-dimethylaminopropyl)carbodiimide; HOBT, 1-hydroxybenzotriazole hydrate; THF, tetrahydrofuran.

mentarity-determining region 2 loop from CD4 behaves in a consistent with the behavior of the intact molecule, in both its binding and kinetic properties. sCD4 is highly sensitive to proteolytic degradation. In contrast, the binding of peptidomimetic CD4-1, like the reovirus mimetic YSGSS (20), is minimally affected by treatment with serum, trypsin, or chymotrypsin (data not shown). In summary, a small conformationally restricted (8-turn mimetic of CD4 residues 40-45 has been designed and synthesized. It effectively inhibits binding of F-gp120 to cell-surface CD4 at low micromolar levels. Similar strategies are being used to explore other important regions of CD4 (33). manner

The ability to readily synthesize a family of conformationally restricted (-turn mimetics that accurately replicate critical elements of peptide secondary structure should prove invaluable for studying a wide range of molecular recognition events.

The CD4 glycoprotein is a member of the immunoglobulin family. Poliovirus and rhinovirus, both members of the picornavirus family, utilize immunoglobulin superfamily molecules as their respective receptors (34-36). Information obtained from this study may provide a general approach to the design of competitive inhibitors of viral receptors belonging to the immunoglobulin superfamily (37). Furthermore, the natural role of immunoglobulin superfamily molecules in growth, cell adhesion, and major histocompatibility supergene

30

-

0

0

X

0

0

0

25-

I-

0

Rc

20-

c

FIG. 3. Stereoview of CD4 residues 40-43 mimetic matched to CD4 residues Gln'4-Phe43, with mimetic in green and CD4 in magenta. There are two low-energy conformers, with the principal structural difference being reversal of the peptide-bond orientation between residues 2 and 3. The energy difference is -0.3 kcal/mol in vacuum; molecular dynamics simulation in a volume-based continuum solution model (22) indicates that there is no significant energy difference between the two conformers under simulated solution conditions. The conformer equivalent to the II' turn found in CD4 is shown here. The six-atom rms deviation between the CD4 atoms and the corresponding mimetic positions a a is

0404A. o.44A.

.

15 0 0

10-II I. I... ... ...,II ... ...,I ... 0 20 40 60 80 100 120

Inhibitor Concentration

(p.g/mn)

FIG. 4. Inhibition of F-gpl20 binding to CD4+ cells, as monitored by reduction in fluorescence intensity. o, CD4-1 mimetic; A, 87.1 reovirus mimetic; *, autofluorescence; *, sCD4. Inhibition by the linear peptide of CD4 residues 40-45 was equivalent to that of the 87.1 reovirus mimetic (data not shown).

Biochemistry: Chen et al.

Proc. Natl. Acad. Sci. USA 89 (1992)

5875

Shen for technical assistance. This work was supported by grants from the National Institutes of Health, the National Science Foundation, a National Science Foundation Presidential Young Investigators Award, a Camille and Henry Dreyfus Foundation Teacher Scholar Award (M.K.), and a grant from the Lucille P. Markey Charitable Trust (M.I.G.). M.K. is an Established Investigator of the American Heart Association.

4540::. 35a)

-D

30-

z

25-

2015

0.0001

0.001

0.01 Concentration (mg/rnl)

1

0.1

44

L-

,-AVj

FIG. 5. Inhibition of syncytia formation by CD4-1 mimetic. (Upper) Number of syncytia per well counted on visual inspection as a function of concentration for CD4-1 mimetic (n), hexapeptide CD4 residues 40-45 (+), and sCD4 (n). (Lower) Photos show syncytia of SupT1 cells cultured with CD4-1 mimetic at 250,ug/ml (A) and CD4 (amino acids 40-45) hexapeptide at 250 Ag/ml (B).

complex restricted antigen presentation (38) could be investigated by this approach. We thank Professor W. C. Still (Columbia University) for a copy of Macromodel, Professor Dan Littmann (University of California, San Francisco) for a preprint of ref. 16, Professor Wayne Hendrickson (Columbia University) for the CD4 coordinates, and Mr. Richard

1. Sattentau, Q. J. & Weiss, R. A. (1988) Cell 52, 631-633. 2. Sleckman, B. P., Peterson, A., Jones, W. K., Foran, J. A., Greenstein, J. L., Seed, B. & Burakoff, S. J. (1987) Nature (London) 328, 351-353. 3. Gay, D., Maddon, P., Sekaly, R., Talle, M. A., Godfrey, M., Long, E., Goldstein, G., Chess, L., Axel, R., Kappler, J. & Marrack, P. (1987) Nature (London) 328, 626-629. 4. Smith, D. H., Byrn, R. A., Marsters, S. A., Gregory, T., Groopman, J. E. & Capon, D. J. (1987) Science 238, 17041707. 5. McDougal, J. S., Kennedy, M. S., Sligh, J. M., Cort, S. P., Mawle, A. & Nicholson, J. K. A. (1986) Science 231, 382-385. 6. Weiner, D. B., Huebner, K., Williams, W. V. & Greene, M. I. (1991) Pathobiology 4, 1-20. 7. Capon, D. J., Chamow, S. M., Mordenti, J., Marsters, S. A., Gregory, T., Mitsuya, H., Byrn, R. A., Lucas, C., Wurm, F. M., Groopman, J. E., Broder, S. & Smith, D. H. (1989) Nature (London) 337, 525-531. 8. Finberg, R. W., Diamond, D. C., Mitchell, D. B., Rosenstein, Y., Soman, G., Norman, T. C., Schreiber, S. L. & Burakoff, S. J. (1990) Science 249, 287-291. 9. Watanabe, M., Chen, Z. W., Tsubota, H., Lord, C. I., Levine, C. G. & Letvin, N. L. (1991) Proc. Nati. Acad. Sci. USA 88, 120-124. 10. Landau, N. R., Warton, M. & Littman, D. R. (1988) Nature (London) 334, 159-162. 11. Ashkenazi, A., Presta, L. G., Marsters, S. A., Camerato, T. R., Rosenthal, K. A., Fendly, B. M. & Capon, D. J. (1990) Proc. Natd. Acad. Sci. USA 87, 7150-7154. 12. Jameson, B. A., Rao, P. E., Kong, L. I., Hahn, B. H., Shaw, G. M., Hood, L. E. & Kent, S. B. H. (1990) Science 240, 1335-1339. 13. Lifson, J. D., Hwang, K. M., Nara, P. L., Fraser, B., Padgett, M., Dunlop, N. M. & Eiden, L. E. (1988) Science 241, 712-716. 14. Wang, J., Yan, Y., Garrett, T. P. J., Liu, J., Rodgers, D. W., Garlick, R. L., Tarr, G. E., Husain, Y., Reinherz, E. L. & Harrison, S. C. (1990) Nature (London) 348, 411-418. 15. Ryu, S.-E., Kwong, P. D., Truneh, A., Porter, T. G., Arthos, J., Rosenberg, M., Dai, X., Xuong, N.-H., Axel, R., Sweet, R. W. & Hendrickson, W. A. (1990) Nature (London) 348, 419-426. 16. Brodsky, M. H., Warton, M., Myers, R. M. & Littmann, D. R. (1990) J. Immunol. 144, 3078-3086. 17. Hruby, V. J., Al-Obeidi, F. & Kazmierski, W. (1990) Biochem. J. 268, 249-262. 18. Marshall, G. R., Gorin, F. A. & Moore, M. L. (1978) Ann. Rep. Med. Chem. 13, 227-238. 19. Veber, D. F. & Freidinger, R. M. (1985) Trends Neurosci. 8, 392-3%. 20. Saragovi, H. U., Fitzpatrick, D., Raktabutr, A., Nakanishi, H., Kahn, M. & Greene, M. I. (1991) Science 253, 792-795. 21. Rose, G. D., Gierasch, L. M. & Smith, J. A. (1985) Adv. Protein Chem. 37, 1-109. 22. Still, W. C. (1990) Macromodel Interactive Modeling System (V3.1X) (Columbia University, New York). 23. Salzmann, T. N., Ratcliff, R. W., Christensen, B. G. & Bouffard, F. A. (1980) J. Am. Chem. Soc. 102, 6161-6163. 24. Williams, R. M., Lee, B. H., Miller, M. M. & Anderson, 0. P. (1989) J. Am. Chem. Soc. 111, 1073-1083. 25. Hoffmann, R. V. & Kim, H.-O. (1990) Tetrahedron Lett. 31, 2953-2956. 26. Tung, R. D., Dhaon, M. K. & Rich, D. H. (1986) J. Org. Chem. 51, 3350-3354. 27. Wasserman, H. H. (1987) Aldrichimica Acta 20, 63-74. 28. Still, W. C., Tempczyk, A., Hawley, R. C. & Hendrickson, T. (1990) J. Am. Chem. Soc. 112, 6127-6129. 29. Cheng, Y. C. & Prusoff, W. H. (1973) Biochem. Pharmacol. 22, 3099-3108.

5876

Proc. Nadl. Acad. Sci. USA 89 (1992)

Biochemistry: Chen et al.

30. Ivey-Hoyle, M., Culp, J. S., Chaikin, M. A., Hellmig, B. D., Matthews, T. J., Sweet, R. W. & Rosenberg, M. (1991) Proc. Natl. Acad. Sci. USA 88, 512-516. 31. Camerini, D. & Seed, B. (1990) Cell 60, 747-754. 32. Moore, J. P., McKeating, J. A., Norton, W. A. & Sattentau, Q. J. (1991) J. Virol. 65, 1133-1140. 33. Peterson, A. & Seed, B. (1988) Cell 54, 65-72. 34. Greve, J. M., Davis, G., Meyer, A. M., Forte, C. P., Yost,

35. 36. 37.

38.

S. C., Marlor, C. W., Kamarck, M. E. & McClelland, A. (1989) Cell 56, 839-847. Staunton, D. E., Merluzzi, V. J., Rothlein, R., Barton, R., Marlin, S. D. & Springer, T. A. (1989) Cell 56, 849-853. Mendelsohn, C. L., Wimmer, E. & Racaniello, V. R. (1989) Cell 56, 855-865. White, J. M. & Littman, D. R. (1989) Cell 56, 725-728. Springer, T. A. (1990) Nature (London) 346, 425-434.